layoutlm-funsd-tf

This model is a fine-tuned version of microsoft/layoutlm-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

• Train Loss: 0.3061
• Validation Loss: 0.6626
• Train Overall Precision: 0.7205
• Train Overall Recall: 0.7863
• Train Overall F1: 0.7519
• Train Overall Accuracy: 0.8006
• Epoch: 7

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• optimizer: {'name': 'AdamWeightDecay', 'learning_rate': 3e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False, 'weight_decay_rate': 0.01}
• training_precision: mixed_float16

Training results

Train Loss	Validation Loss	Train Overall Precision	Train Overall Recall	Train Overall F1	Train Overall Accuracy	Epoch
1.6700	1.3895	0.2812	0.2880	0.2846	0.5333	0
1.1465	0.8828	0.5819	0.6433	0.6111	0.7324	1
0.7878	0.7302	0.6497	0.7240	0.6849	0.7791	2
0.6195	0.6630	0.6844	0.7692	0.7243	0.7949	3
0.5031	0.6265	0.6929	0.7903	0.7384	0.8112	4
0.4202	0.6244	0.7329	0.7863	0.7587	0.8124	5
0.3479	0.6233	0.7275	0.7702	0.7482	0.8127	6
0.3061	0.6626	0.7205	0.7863	0.7519	0.8006	7

Framework versions

• Transformers 4.25.1
• TensorFlow 2.9.0
• Datasets 2.8.0
• Tokenizers 0.13.2